Development of High-Strength Nitriding Steel for Gear Hideki

NIPPON STEEL & SUMITOMO METAL TECHNICAL REPORT No. 116 September 2017

Technical Report UDC 669 . 14 - 422 . 11 : 669 . 14 - 155 : 621 . 833 Development of High-strength Nitriding Steel for Gear

Hideki IMATAKA* Masato YUYA Kosuke TANAKA Atsushi KOBAYASHI Susumu MAEDA

Abstract Transmission gears require high fatigue strength and low heat-treatment distortion for high performance. We focused on the gas-nitriding process that was conducted at a transformation temperature lower than that of martensite and suppressed distortion and developed high-strength steel designated for nitrided gears. High surface hardness and deep case depth was achieved by optimizing the content of Cr and V in low carbon steel. The results of rotary bending and the roller pitting fatigue tests showed the fatigue strength of developed steel was equal to or higher than that of conventional carburized steel. Observation by transmission electron microscopy showed fine carbonitrides including both Cr and V pre- cipitated at the nitride layer of developed steel. This suggests that fine precipitates contribute to optimal hardness distribution and excellent fatigue properties.

1. Introduction through carburizing and quenching processes, the gears are hard-

Higher fuel efficiency and less CO2 emission have become in- ened by having carbon penetrate into the material from the surface creasingly required of road vehicles recently in consideration of and diffuse to inside. During the treatment, however, the steel is conservation of the global environment. Accordingly, the size and heated to the austenitic temperature or higher, incurring the problem weight of transmission gears that compose the automotive power of large distortion throughout the process. With the ring gears for train have been significantly reduced through various measures. 1–4) planetary systems, especially, it is difficult to obtain the required di- Carburizing and quenching, 5) widely employed to strengthen gears, mensional accuracy by the carburizing and quenching process be- incurs large distortion during the treatment because the process in- cause of their large diameter and thin thickness. Moreover, because volves structural phase transformation, and the gear surfaces have to of the internal teeth, their re-machining after the treatment is prob- be ground after the treatment to reduce gear noise, which incurs ad- lematic. Therefore, to reduce the heat treatment distortion, there ditional costs. For this reason, both high fatigue strength and low have been many attempts to develop new steels for nitriding, 9–14) a distortion must be realized at low cost. To this end, we focused on surface hardening process at a lower temperature. It was difficult nitriding, 6–8) which is conducted at a comparatively low temperature with nitriding steel, however, to obtain a fatigue strength equal to or and causes less distortion, and successfully developed a new high- higher than that of carburized and quenched steel and a material strength nitriding steel for gears. hardness adequate for gear machining at the same time. The present paper explains the design philosophy of the chemi- In the present development, we aimed at obtaining high-strength cal composition of the developed steel, and explains the studies on nitriding steel for gears 15) having a hardness adequate for the ma- its hardening mechanism, heat treatment distortion and fatigue prop- chining before the treatment and the fatigue strength equivalent to erties. that of carburized and quenched steel after it. At the same time, regarding the distortion due to nitriding, we focused on the volume 2. Development Target expansion of the nitrided layer as a likely cause, and studied the in- Automotive transmission gears are mostly made of carburized fluences of alloy elements over steel expansion. and quenched alloy steel for machine structural use in appreciation of high resistance to fatigue and wear and good machinability. The 3. Improvement of Steel Strength steel is cut into gears before hardening heat treatment, and then In the present development study, the steel chemistry of JIS

* Senior Manager, Bar & Wire Rod Quality Control Dept.-I, Bar & Wire Rod Div., Yawata Works 1 Konomi-machi, Kokurakita-ku, Kitakyushu City, Fukuoka Pref. 803-0803

- 9 - NIPPON STEEL & SUMITOMO METAL TECHNICAL REPORT No. 116 September 2017

SCr420H, a popular alloy steel for machine structural use, was ished by buffing before the measurement. modified to obtain good machinability for gear cutting before nitrid- [{d (1) + d (2) + ... + d (16)} −16 × 200] / 16 (1) ing as well as high fatigue strength after the treatment. Table 1 Fatigue strength was evaluated by the Ono-type rotating bending shows the chemical compositions of related steels; here JIS S35C, a fatigue test and roller pitting test using test pieces after nitriding. popular carbon steel for machine structural use, is referred to as a The former test was conducted at room temperature at a rotation of comparative steel. 3 000 rpm using V-notched test pieces with a stress concentration ra- First, regarding nitriding properties, we considered the addition tio of α = 2.0, and the fatigue strength was the maximum stress that of Cr and V, which form nitrides in the surface layer during nitriding the test piece could withstand for 107 cycles of bending. The latter to increase surface hardness. Next, to improve nitriding properties test was conducted at a rotation of 1 500 rpm using an automatic lu- and lower the material hardness before the treatment, we aimed at bricant oil with a slipping ratio of −40% at 90°C, and the fatigue lowering carbon content to increase the percentage fraction of fer- strength was the maximum stress that the test piece could withstand rite. In addition, the contents of other alloy elements were set such for 2 × 107 cycles of pitting. that a ferrite-pearlite structure not containing bainite would be ob- tained for good machinability. 5. Test Results 5.1 Machinability 4. Tests Machinability was evaluated by a turning test. The specimens of The developed steel and S35C were prepared using a converter, substantially the same hardness were used: the developed steel HRB continuously cast into blooms, rolled into billets and then into bars. 86, and S35C HRB 85. As seen in Fig. 2, the turning force was The bars were then machined into test pieces, and subjected to a nearly the same with both steels. The turning chips are shown in turning test before nitriding. Then, they underwent gas nitriding at Fig. 3; the chips of both specimens were divided into short pieces, 600°C for 2 h, and their properties were examined. The sectional evidencing good machinability of the two steels. hardness distribution after nitriding was measured using a micro- 5.2 Nitriding properties Vickers hardness tester. The microstructure of the surface layer was Figure 4 compares the two steels in terms of sectional hardness observed through an optical microscope, and precipitations in the distribution after nitriding. The developed steel, which contains Cr layer were observed through a scanning electron microscope (SEM) and V, had far higher hardness at the surface layer than that of S35C. and a transmission electron microscope (TEM). Considering that the surface hardness of common carburized steel is Distribution of nitrogen concentration was measured using turn- mostly in the range of HV 700 to 800, the developed steel has the ing chips cut out from the surface layer. Because the distortion (vol- same surface layer hardness as that of carburized steel. ume expansion) due to nitriding is very small, it was measured by Figure 5 shows the measured nitrogen concentration distribution the method given in Fig. 1, that is, depression marks were pressed at in the surface layer after nitriding. The points represent the analysis equal intervals of 200 μm using a micro-Vickers hardness tester on results of the turning chips taken at intervals of 50 μm from the sur- one surface of the test pieces of square bars, each 10 × 10 × 20 mm in face. The nitrogen concentration in the surface layer of the devel- size, the change of the intervals before and after the nitriding was oped steel was more than twice that of the comparative steel, S35C. measured through the SEM, and the expansion was evaluated using This is presumably because the developed steel contains Cr and V, the value of (1) below as the indicator. Note that, when the press which have an affinity for N. marks were unclear after the treatment, the surface was lightly pol- Figure 6 shows the microstructures in the surface layer of both the specimens after nitriding. While the structure of the two speci- Table 1 Chemical compositions (mass%) mens was ferrite + pearlite, the developed steel, which contains less carbon, had a higher percentage fraction of ferrite than that of S35C. Material C Si Mn Cr V Note that both the specimens had a layer of compounds, roughly 5 Developed μm in thickness, at the outermost layer. 0.10 0.15 0.55 1.25 0.17 high-strength steel JIS SCr420H 0.20 0.20 0.80 1.00 – JIS S35C 0.35 0.20 0.75 – – (conventional steel)

Fig. 1 Method of measuring distortion Fig. 2 Turning force Fig. 3 Photos of chips

- 10 - NIPPON STEEL & SUMITOMO METAL TECHNICAL REPORT No. 116 September 2017

Figure 7 shows the microstructures of the specimens at 50 μm from the fact that all the precipitates in the developed steel are of the from the surface after nitriding observed through the SEM. Whereas same shape and size, they are presumed to be complex carbo-ni- there were many acicular crystals of iron nitride (marked with ar- trides of Cr and V, or (Cr, V) (C, N). rows) in the ferrite of the S35C specimen, few similar types were Next, Fig. 10 shows the change in the hardness of surface layers found in the developed steel. after nitriding of specimens prepared based on the developed steel Figure 8 shows TEM photomicrographs of the specimens at 50 and changing the contents of the alloying elements (C, Mn, Cr and μm from the surface after nitriding; fine precipitates were seen to V) individually. Decreasing C and increasing Cr and V are especial- have formed in the nitrided layers of the two steels. Considering ly effective at increasing the surface layer hardness. their distribution, stretched grains, some tens of nanometers in length, were sparsely observed in S35C, and in the developed steel in contrast, finer grains, several nanometers in size, were seen more densely. The difference in the precipitate distribution corresponded well to that of the surface hardness, which indicates that the higher hardness of the nitrided layer of the developed steel than that of S35C is due to the difference in the distribution of the fine precipitates. From the selected-area diffraction pattern of a precipitate of the developed steel given in Fig. 9, it proved to be a FCC crystal in the Baker-Nutting orientation relationship with the parent phase. Ei- ther Cr or V is capable of forming FCC carbo-nitrides, but judging Fig. 6 Microstructures of nitride layers near surface after nitriding (nital etching)

Fig. 7 SEM images of nitride layers near the surface (nital etching)

Fig. 4 Hardness profiles of nitrided layers of developed steel and S35C

Fig. 8 TEM dark field images of precipitates taken at a depth of 50 μm below the surface

Fig. 5 Nitrogen concentration profiles at the nitrided layers of devel- Fig. 9 Selected area diffraction pattern taken in nitrided developed oped steel and S35C steel and corresponding key diagram

- 11 - NIPPON STEEL & SUMITOMO METAL TECHNICAL REPORT No. 116 September 2017

Fig. 10 Influence of alloying elements on surface hardness after nitriding

Fig. 12 Results of Ono-type bending fatigue test

Fig. 11 Expansion due to nitriding Fig. 13 Results of roller pitting fatigue test 5.3 Deformation due to nitriding treatment Figure 11 shows the evaluation result of the distortion (volume are in ratios to the fatigue strength of gas nitrided S35C after 107 cy- expansion) due to gas nitriding of the steels prepared by changing cles of bending, given as 1.0. The developed steel proved to have a the contents of the alloying elements (Mn, Cr, Mo and V) individu- bending fatigue strength 3.2 times larger than that of S35C, and sub- ally from those of the developed steel. Using multiple regression stantially equal to that of gas carburized JIS SCr420H, as indicated analysis, the volume expansion could be expressed as given in (2) by the dotted curve. below; the calculated figures agreed well with the measurements. Figure 13 shows the result of the latter test. Here, as in Fig. 12, 0.61 Mn + 1.11 Cr + 0.35 Mo + 0.47 V (2) the points are in ratios to the fatigue strength of gas nitrided S35C Here, Mn, Cr, Mo and V represent the contents of the respective ele- after 2 × 107 cycles of bending, defined as 1.0. The developed steel ments. From the above, it is clarified that each of their contents has had an anti-pitting strength higher than that of S35C by 28%, and positive effects over the expansion, and the more a steel contains substantially equal to that of SCr420H after gas carburizing, shown these elements, the more it expands through nitriding. by the dotted curve. All the above results indicate that the developed steel has an original hardness suitable for gear machining and excellent nitriding 6. Conclusion properties when designed to contain appropriate amounts of Cr and A new low-distortion, high-strength nitriding steel for gears hav- V on the basis of low-C steel. In addition, minimizing the expansion ing significantly higher strength than conventional steels was devel- due to gas nitriding has been enabled by making the value of (2) as oped. The development studies revealed the following findings: small as possible within the content ranges of the alloy elements to (1) To improve the surface hardness after nitriding, it is effective satisfy the required properties. to add appropriate amounts of Cr and V to low-C steel. 5.4 Fatigue strength (2) Fine precipitates of the carbo-nitrides of the alloying elements, Using nitrided specimens of the developed steel and S35C, an (Cr, V) (C, N), which form in quantities in the surface layer of Ono-type rotating bending fatigue test simulating the fatigue at gear the developed steel during nitriding, serve to increase surface tooth roots and roller pitting test simulating that at tooth faces were hardness. conducted. (3) The volume expansion due to nitriding, which can be ex- Figure 12 shows the result of the former test. Here, the points pressed as a function of the contents of Mn, Cr, Mo and V, can

- 12 - NIPPON STEEL & SUMITOMO METAL TECHNICAL REPORT No. 116 September 2017

be minimized by decreasing the contents of these elements Jpn. Soc. Heat Treat.). 32 (5), 262–266 (1992) 8) Ohki, T.: Trend of Nitriding Steel. Netsu Shori (J. Jpn. Soc. Heat Treat.). within the content ranges to satisfy the properties required for 30 (6), (1990) the steel. 9) Kanbara, S., Aihara, K., Okuyama, S.: Effect of Carbon and Lead on (4) The developed steel has a bending fatigue strength 3.2 times Ferritic Nitro-carburizing Properties of Cr-V Steels. Netsu Shori (J. Jpn. that of S35C, and an anti-pitting strength higher than that of Soc. Heat Treat.). 26 (5), 339–344 (1986) 10) Izumi, K., Kamada, Y., Takayama, T., Hinotani, S. et al.: Effects of Soft- S35C by 28%. These figures are substantially the same as those Nitriding Conditions on Pitting Characteristics of SoftNitriding Steel. of gas carburized SCr420H. Proc. Jpn. Soc. Heat Treat. 39, 19–20 (1994) 11) Kobayashi, M., Sakurada, T., Okabe, I.: Prediction Method of Fatigue References Strength of Gas Soft-Nitrided and Ion Nitrided Steels. Trans. Jpn. Soc. 1) Iwama, N.: Trend of Special Steel for Automotive Parts. Jidosha Gijutu (J. Mech. Eng. Series A. 62 (597), 1132–1139 (1996) Soc. Automotive Engineers Jpn.). 55 (10), 64–68 (2001) 12) Ishikawa, N., Shiraga, T., Ishiguro, M. et al.: Pitting Fatigue Behavior of 2) Hayashi, T., Kino, N., Okada, Y.: Material Technology for Downsizing Gas Nitrided Spur Gear (Development of High Strength Nitriding Steel and Weight Reduction of DriveTrain. Nissan Tech. Review. (57), 43–47 for Gears - 5). CAMP-ISIJ. 10 (3), 474 (1997) (2005) 13) Inoue, K., Matsumura, Y.: Influence of Alloying Elements on Hardness 3) Kino, N., Mabuchi, Y., Koyanagi, K., Murakami, R.: Technical Review Distribution after Nitrocarburizing in Medium Carbon Steel. Denki Sei- of Materials for Automotive Power Train. Denki Seiko (Electr. Furn. ko (Electr. Furn. Steel). 75 (1), 11–18 (2004) Steel). 85 (2), 97–106 (2014) 14) Masuda, H., Matsuda, H., Yamazaki, A.: Development of High-strength 4) Satoh, N.: Latest Trend of Special Steel for Automotive Use and Materi- and High-wear Resistance Nitrocarburizing Steel by Precipitation al Development. Sanyo Technical Report. 8 (1), 68–87 (2001) Strengthening of Mo-V Carbonitride. Honda R&D Technical Review. 24 5) Sumida, M., Nomura, I.: Trend and Challenge in Carburized Forged (2), 111–119 (2012) Parts of DriveTtrain. Netsu Shori (J. Jpn. Soc. Heat Treat.). 45 (2), 76– 15) Kobayashi, A., Maeda, S., Imataka, H., Gyotoku, Y., Yuya, M., Shimizu, 79 (2005) Y., Kanayama, M.: High-strength Nitriding Gear Developed by γ'-Fe4N 6) Takase, T.: Surface Hardening of Steel by Nitriding Method. Tetsu-to- Phase and Low Carbon Alloy Steel. Proc. Soc. Automotive Engineers Hagané. 66, 1423–1434 (1980) Jpn. (JSAE), 2014. No. 14–14, 21–26 (2014) 7) Suzuki, N., Tani, K.: Recent Trend of Nitrided Products. Netsu Shori (J.

Hideki IMATAKA Atsushi KOBAYASHI Senior Manager HONDA R&D Asia Pacific Co., Ltd. Bar & Wire Rod Quality Control Dept.-I Bar & Wire Rod Div., Yawata Works 1 Konomi-machi, Kokurakita-ku, Kitakyushu City, Fukuoka Pref. 803-0803 Masato YUYA Susumu MAEDA Senior Researcher Honda R&D Co.,Ltd. Bar & Wire Rod Research Lab. Steel Research Laboratories

Kosuke TANAKA Bar & Wire Rod Quality Control Dept.-I Bar & Wire Rod Div., Yawata Works

- 13 -